Christopher Delgado Electrical Engineering
Emmanuel Hernandez Electrical Engineering
Jason Balog Electrical Engineering
Ramon Santana Electrical Engineering
Group 24
Sign Language Interpreter Glove
Thank You to our generous sponsors!!!
Motivation
● The Survey of Income and Program Participation
(SIPP) – estimates that about 1,000,000 are
functionally deaf in the United States.
● The World Health Organization estimates that over
5% of the world‟s population – 360 million people –
has disabling hearing loss (328 million adults and
32 million children).
● The original motivation to pursue this project comes
from one of our team members who has
experienced the difficulty of communicating with his
speech-impaired sister.
Goals and Objectives
● Our objective is to establish communication between a sign language speaker and a non-sign language
speaker. Any letter the user signs will be displayed through a user interface where the non ASL-speaker can
read the letter. We also want to implement a learning mode, where the user has the option to learn the
American Sign Language letters.
● Hardware
● Flex sensors
● Contact sensors
● Accelerometer
● Gyroscope
● MCU: ATMega328
● Bluetooth Low Energy
● Software
● Android Mobile Application
Specifications
● Lightweight
● Portable
● Energy Efficient
Glove Weight < 1.5 lb.
Lithium Ion Battery 3.7 V 2000mAh
Battery Life 13 hours
Specifications Table
Related Standards
● IEEE Standards 802.15.1™-2005(Revision of IEEE Std 802.15.1-2002)
-Establishes a communication standard optimized to serve a variety of medical and non-medical
applications.
● Safety standards for Lithium-ion batteries (International Electrotechnical Commission (IEC) and
International Organization for Standardization (ISO).
-IEC 62133-2: safety requirement for portable battery cells
-IEC 62660: batteries for EV/HEV applications
-IEC 61427: secondary cells and batteries for renewable energy storage
● Google's Android developers have set up a multitude of standards and qualification on the different
aspects of an android application.
-UX-B1: App uses common user interface patterns and conventional use of icons
Block Diagram
Significant Hardware Decisions - Wireless Communication
Bluetooth Low Energy NFC
Pros:
- Security (WPA2)
- Range 150-300 feet
Pros:
- Range 20-120 feet
- Consumes less
power than Wi-Fi
Pros:
- Consumes less
power than Bluetooth Low
Energy
Cons:
- Needs Router
- Consumes lots of
Power
- Not portable friendly
Cons:
- No cons for SLIG ☺
Cons:
- Limited Range
(about 0-4 inches)
Significant Hardware Decisions - Battery & Regulators
● Polymer Lithium Ion Battery - 2000mAh
● 3.7V at 2000mAh
● Built-in protection against over voltage
and over current
● Self-discharge rate <8% per month
● Regulators - TI LP2985
● The purpose of voltage regulators is to
keep a constant voltage level
● Dropout Voltage: 0.4 V
● Output Voltage: 3.3 V
Significant Hardware Decisions - Charging
● Charger input:
● Micro-USB
● Charge current 500mA
● Charger output:
● Single-Cell batteries Only
● Lithium-Polymer or Lithium-Ion Only
● Max Voltage: 4.2 V
● Top challenges were:
● Learning how to solder small electronics
Significant Hardware Decisions – Flex Sensors
● Flex sensors will be the primary sensors employed in
this projects.
● Flex sensors will be used to detect the degree to
which each finger is bent on the hand performing the
sign language gesture.
● Each letter will have a specific configuration based
on expected outputs of the five flex sensors.
Significant Hardware Decisions – Flex Sensors
SpectraSymbol Long Flex
Sensor
Images Two-Directional Bi-
Flex Sensor Tactilus Flex
Pros:
- widely available
- accurate output
Pros:
- built in pressure sensors
Pros:
- claims high durability
(>35 million cycles)
- high quality leads
Cons:
- fragile leads
Cons:
- irregular shape
- sold exclusively by
manufacturer
Cons:
- increased cost
- sold exclusively by
manufacturer
Significant Hardware Decisions – Gyroscope and
Accelerometer
● Accelerometers and gyroscopes can be used
to measure these type of parameters, which
are crucial in identifying certain sign language
letters.
● Examples include "j" and "z" or distinguishing
between “g” and “q”.
● The group is using the SparkFun 9 Degrees of
Freedom IMU Breakout - LSM9DS1.
Significant Hardware Decisions – Contact Sensors
● Contact sensors were crucial in telling apart
the following pairs of sign language: R, U,
and V, S and T and M and N.
● The contact sensors were made with strips
of copper braids that were connected to the
power supply through wire leads.
● The glove detects when two contact sensors
come in contact creating a closed circuit.
Significant Hardware Decisions – PCB
● The PCB was designed using Eagle CAD software and
fabricated by Bay Area Circuits and most parts were sourced
from Digikey.com
● It was designed as two layer board measuring 2.5” by 3.5”
and is meant to be worn on the forearm.
● Designing and building the PCB board was the biggest
challenge with the hardware.
● Top challenges were:
- Learning Eagle CAD
- Correctly Wiring Components
- Ordering the write components
Significant Hardware Decisions – PCB
Significant Hardware Decisions – Microcontroller
● The unit that is chosen will have to employ at
least 6 analog input pins (with included ADC
units), have digital i/o pins, and also be able to
establish a line of serial communication.
● When considering the microcontroller unit, it is
also important to note that the two units come
with their own native programming
environments.
● The gesture recognition system will
use thresholds to determine what
hand gesture is being performed.
● The ranges of values that are
included in every type of hand
gesture determine the minimum and
maximum limits to the conditional
statements.
Gesture Recognition Software
Calibration Process
● The calibration process occurs during the „setup‟
process of the program, before the infinite loop
begins
● Calibration will capture the minimum and maximum
input values from each sensor, and apply the
mapping function.
● Calibration is needed because analog flex sensors
provide volatile data that needs to be normalized.
Gesture Recognition Software
● Program scans sensors for data at one second intervals
● Arduino “map” function is used to apply calibration
● Normalized sensor values and motion data are used in
conditional statements that determine gesture being
performed
Gesture Recognition Set-Backs
● Contact sensors needed to have internal pull-up
resistors enabled due to floating i/o pins
● Had to use SPI as opposed to I2C for the motion
unit due to a lack of enough analog input pins
● Many distinct gestures unexpectedly produced
similar flex sensor data and required methods of
distinction
Software Components: Mobile Application
Mobile Application: Overview
● The mobile application is an important feature of this
project that will serve as the user interface for the
sign language glove.
● The mobile application is responsible for wirelessly
displaying hand gestures performed by the glove onto
a mobile phone screen as text.
● In order for the mobile application to be successful,
the design of the app should consider the user, be
simple and elegant, and meet all design
requirements.
Potential Mobile Platforms
● Android ● Programming language: Java ● IDE: Android Studio
● Open source and cross-platform compatible
● BLE compatible ● Everyone in our group owns one!
● iOS
● Programming Languages: Swift or Objective-C ● IDE: Xcode
● Only Mac OS compatible
● BLE compatible
● Windows ● Programming Languages: C++, C#, and Visual Basic ● IDE: Visual Studio ● BLE compatible
Android
Minimum Requirements
Device Type Android
Mobile Platform Smartphone
Bluetooth Version Bluetooth Low Energy
v4.0
Platform Version Android 4.3
Codename Jelly Bean-MR2
Android API Level 18
Mobile App
Flowchart
User Interface
User Interface
Budget And Financing
● Sponsored by: Boeing and Leidos
● Initial Budget Estimate: $550
● Final Project Cost: $825
Part Description Price per Unit ($) Quantity Cost ($)
Flex sensors 4.5" $12.95 15 $194.25
Flex sensors 2.2" $7.95 1 $7.95
Development Board $20.87 3 $62.61
ATmega328P $13.99 3 $41.97
Bluetooth Module $11.29 4 $45.16
Accelerometer & Gyroscope $29.55 2 $59.10
Battery $22.20 1 $22.20
PCB $49.60 1 $49.60
PCB Components $49.76 1 $49.76
2nd PCB Components $52.99 1 $52.99
3rd PCB Components $34.08 1 $34.08
Glove and Wristband $20.22 2 $40.44
Glove and Armband $45.00 1 $45.00
Soldering Materials et al $120.00 1 $120.00
Total cost - - 825.11
Boeing & Leidos sponsorship - - $359.80
Difference $465.31
Questions ?
#SLIG
Top Related